Digital Agriculture System for Crop Prediction Based on Machine Learning

: Agriculture is the number one source of livelihood for approximately 58 percent of India’s population and is the most crucial part of GDP. Indian farming is based totally on economic advantages from crop yields, but now day’s agricultural generation has failed to verified satisfactory crop choice techniques and to boom crop yield in all over India. So, lower in crop yield will increase the trouble in farmer’s monetary health situations. So, it will become the maximum trending hassle for our agricultural region to invent such noble technique to advocate super appropriate crop for a particular region. To reap high-quality appropriate crop desire for areas primarily based on parameters like soil conditions, rainfall and weather we have got applied gadget studying method. Secondary hassle is lack of understanding or absence of steering even as farming. Lack of guidance in Indian farmers may follow incorrect farming strategies or inefficient traditional strategies. Most of farmers are uneducated and non-technical backgrounds so they'll be relying on conventional crop choice and farming techniques which falls them into reasonable loss. With the assist of disorder assessment tool, we predict the crop disease prediction and propose the precaution from the ones illnesses. Last and most essential hassle isn't any right marketplace assessment at the equal time as cultivation of any unique crop, which can also reason a cheap lack of farmers.


I. INTRODUCTION
Indian farming is based on financial advantages from crop yields, but now day's agricultural technology has didn't demonstrated high-quality crop selection strategies and to growth crop yield in throughout India.So lower in crop yield increases trouble in farmer's economic fitness conditions.So it becomes most trending hassle for our agricultural area to invent such noble approach to suggest excellent appropriate crop and need of on-line market place.Crop recommendation is absolutely based totally on environmental factors like soil, weather and rainfall for particular region.So there's want of device mastering techniques like guide vector gadget and Convolutional neural community for classification and clustering dataset.We recommend excellent suitable crop for precise location based totally in this nearby parametric environmental facts.Our contribution will solves crop choice hassle and ultimately boom the price of yields and allows improving monetary health of our farmers.As properly as we are going to provide guidance by technical strategies from cultivation to yields by way of helping out dynamic queries of farmers whilst farming.We are going to provide E-mart for on line promoting of farmer's product that allows you to help to farmers to get more little expensive benefits than present marketplace charge.Miss. Snehal S. Dahikar,Dr. Sandeep V.Rode et. al.[1], introducing by considering completely different circumstances of climatologically marvels influencing neighbourhood climate conditions in several items of the planet.These climate conditions directly have an effect on harvest yield.Completely different examines are finished investigation the associations between scale climatologically wonders and harvest yield.Counterfeit neural systems are shown to be integral assets for demonstrating and forecast, to make their viability.Saeed Khaki and Lizhi Wang et. al.,[2], proposed that crop yield is associate degree exceptionally puzzling quality determined by various parts, for instance, genotype, condition, and their cooperation's.Precise yield expectation needs essential comprehension of the helpful association among yield and these intuitive variables, and to uncover such relationship needs each thorough datasets and wonderful calculations.

II. LITERATURE SURVEY
Rakesh Kumar et.al.
[3], Proposed agriculture composing assumes an enormous job in monetary development and nourishment security of agro-based nation.choice of crop(s) may be a important issue for business enterprise composing.It depends upon completely different parameters, for instance, generation rate, and showcase price and government arrangements.varied analysts contemplated expectation of yield pace of harvest, forecast of climate, soil order and harvest arrangement for husbandry composing utilizing measurements methods or AI systems.
Pritam Bose et.al[4], state that Convolutional neural networks (CNNs) for remote detection spatiotemporal investigation of image time arrangement that utilize the exceptionally parallel and low-control avid xenomorphic hardware platforms doable.This technique delineates this concept with the presentation of the first CNN process model for harvest yield estimation from standardized distinction vegetation record image time arrangement.It introduces the advance and testing of a method structure that uses the spatial accumulation of your time series of Moderate Resolution Imaging Spectradiameter 250-m resolution knowledge and historical crop yield knowledge to organize a CNN to form auspicious forecast of harvest yield.
Francisco Yandun et.al [5] ,states that Farming advent have to twofold by way of 2050 so one can fulfill the normal nourishment need because of populace improvement.Exactness horticulture is the way to improve profitability and efficiency within the usage of belongings, eventually engaging in this objective under the extraordinary difficulties as of now regarded by way of agribusiness mainly because of ecosystem modifications, land corruption, accessibility of farmable land, paintings energy lacks and expanding prices.To confront these difficulties, accuracy agribusiness uses and creates detecting structures that supply information approximately the harvest improvement and wellbeing guidelines.
Prof. K. A. Patil and Prof. N. R. Kale et.al., introducing Atmosphere changes and precipitation has been capricious over the previous decade.Attributable to this in late time, atmosphere good techniques known as as shrewd business enterprise is received by varied Indian ranchers.good farming may be a robotized and coordinated information innovation dead with the IOT (Internet of Things).IOT is growing quickly and loosely applied in each single remote condition.

III. OBJECTIVE AND GOALS
Agriculture is the main base of Indian economy.The agriculture era is the most important economical sector in our county.The farmers are totally depends on the crops and their farms for economical gain.The yield gets based on weather conditions as rainfall structure highly influences cultivation steps.So there is need of farmers and agriculturalists require spontaneous advice proposition in predicting future reaping instances to maximize crop yield.
We have to collect all required data by giving GPS locations of a land and by taking access from Rain forecasting system of by the government, we can predict crops by just giving GPS location.Also, we can develop the model to avoid over and under crisis of the food.

IV. EXISTING SYSTEM APPROACH
Agriculture is the principle base of Indian financial system.In India, farmer used crop selection method is only conventional technique.The agriculture technology is the most vital and powerful financial quarter in our county.The farmers are absolutely relying on the vegetation and their farms for least expensive gain.The yield obtained generally relies upon on weather situations as rainfall patterns largely influence cultivation methodologies.So, want of farmers and agriculturalists require a spontaneous recommendation proposition in predicting future reaping times to maximise crop yield.In traditional manner on machine gaining knowledge of and agriculture analysis we came throughout the truth that traditionally crop choice techniques is not pleasing the farmers cost effective delight.We are confronted such a lot of issues in present paintings.Due to incorrect or flawed crop choice approach GDP is likewise low.

V. METHODOLOGY 5.1 Dataset Gathering
There are two data sets used for the our model.The first contains historic district-wise rainfall data for Pune districts of Maharashtra

A. Climate and Rainfall
At the Western Ghat and hill region is cool and eastern region having hot and dry climate.The maximum temperature of Pune district ranges between 34 and 410C during April-May, while the minimum temperature varies between 50C to 100C in the months of November to January.The average annual rainfall at the district is 675 mm, most of which is receive during South-West monsoon.However, medium rainfall zone at district having on average rainfall of 900 mm, eastern region have an average between 600 to 700 mm while western region have an average of 1171 mm5.The regularity in occurrence in recent years has not experienced in the district.

B. Soil and Topography
Pune district possesses mainly three types of soils, viz.black-fertile, brown and mixed type.In western region soil, type has brown and low quality while eastern region having fertile and plain type.The richest alluvial soil track found in the Valley of Bheema River.The rivers Velu, Ghod are left side of Bheema and Indrayani, Bhama, Mula-Mutha etc. are at right side.Each tahsil of the district have minimum one river6.Therefore, the agro-climatic condition of district is favourable.
 Convolutional Neural Network (CNN): A Convolutional Neural Network is a Deep Learning algorithm which can take in an input image, assign importance to various aspects/objects in the image and be able to differentiate one from the other.It is a class of artificial neural networks that has become dominant in various computer vision tasks, is attracting interest across a variety of domains, including radiology.CNN is designed to automatically learn spatial hierarchies of features through backpropagation by using multiple building blocks, such as convolution layers, pooling layers, and fully connected layers.The first two layers convolution and pooling perform feature extraction whereas the third layer a fully connected maps the extracted features into final output, such as classification. Machine Learning: Machine learning is an application of artificial intelligence (AI) that provides systems the ability to automatically learn and improve from experience without being explicitly programmed.Machine learning focuses on the development of computer programs that can access data and use it to learn for themselves  Image Processing: Image processing is a method to perform some operations on an image, in order to get an enhanced image or to extract some useful information from it.It is a type of signal processing in which input is an image and output may be image or characteristics/features associated with that image.
VI. PROPOSED SYSTEM APPROACH Agriculture is the critical issue of economy in India.In current years because of industrialization; excessive use of insecticides the electricity of soil is getting affected.Many of the methods observed through agriculture aren't enough to growth the productivity.The commonplace problem gift the various Indian farmers are they don't have any data regarding the right crop primarily based on their soil requirements so it impacts the productivity.Thus, we try to prove the current crop selection technique influences on farmers within your means ability by using degrading yield boom.So, we invent the effective crop choice technique primarily based on machine learning (SVM).We advise the first-class appropriate crop for the regions thinking about environmental conditions.Agriculture is the backbone for a growing economy like India and there is a sizable need to preserve the rural sustainability.We are going to offer one solution for all make our system smart and virtual vicinity for agriculture.1: Modules of System We proposed assistive system for economical welfare of farmers.The existing problem is economical down condition of farmers due to wrong crop cultivation, lack of current market sense, lack of guidance and market place unavailability.The farmers select crops based on traditional methods.There is not any current market evaluation consider while cultivation of crops.There is also lack of guidance to farmers from cultivation of crops to uprooting of crops.And the last one most important fix market place unavailability.
We are creating support vector machine algorithm for classifying environmental factors.The environmental factors are like soil types with NPK values, 10 years rainfall dataset, and temperature dataset.The trained model is constructed based . The collection period spans to 10 years from 2010 to 2018.Rainfall is measured in millimetres and the labelled volume for a District is the mean of values recorded at all the weather stations in the District.The other data set contains a detailed description about the soil properties recorded in Pune District of Maharashtra recorded over 10 years.Soil properties include the concentration of Nitrogen, Phosphorous and Potassium (NPK) in the soil (all in tones), the scales of pH of the soil, amongst others.Every row of values is labelled with a corresponding Yield value expressed in tones per hectare.The trained model proposed in this paper curates results of the model trained on rainfall data with the machine learning model trained on other soil properties.

Figure 1 :
Figure 1: Block Diagram of Proposed System In the proposed work we are going to invent: In proposed system consist of 5 modules:- Admin: In this module, the Admin has to log in by using valid user name and password.After login successful he can do some operations such as View All Users and Authorize, View All E-Commerce Website and Authorize, View All Products and Reviews, View All Products Early Reviews, View All Keyword Search Details, View All Products Search Ratio, View All Keyword Search Results, View All Product Review Rank Results.Register, login, view farmer's details, view all crop details, update crop details and monitor system. View and Authorize Users: In this module, the admin can view the list of users who all registered.In this, the admin can view the user's details such as, user name, email, address and admin authorizes the users. View Chart Results: View All Products Search Ratio, View All Keyword Search Results, View All Product Review Rank Results. Ecommerce User: In this module, there are n numbers of users are present.User should register before doing anyoperations.Once user registers, their details will be stored to the database.After registration successful, he has to login by using authorized user name and password Once Login is successful user will do some operations like Add Products, View All Products with reviews, View All Early Product's reviews, View All Purchased Transactions. End User: In this module, there are n numbers of users are present.User should register before doing any operations.Once user registers, their details will best or to the database.After registration successful, he has to login by using authorized user name and password.Once Login is successful user will do some operations like Manage Account, Search Products by keyword and Purchase, View Your Search Transactions, View.
After login successful he can do some operations such as View All Users and Authorize, View All E-Commerce Website and Authorize, View All Products and Reviews, View All Products Early Reviews, View All Keyword Search Details, View All Products Search Ratio, View All Keyword Search Results, View All Product Review Rank Results.Register, login, view farmer's details, view all crop details, update crop details and monitor system. View and Authorize Users: In this module, the admin can view the list of users who all registered.In this, the admin can view the user's details such as, user name, email, address and admin authorizes the users. View Chart Results: View All Products Search Ratio, View All Keyword Search Results, View All Product Review Rank Results.
 Ecommerce User: In this module, there are n numbers of users are present.User should register before doing any operations.Once user registers, their details will be stored to the database.After registration successful, he has to login by using authorized user name and password Once Login is successful user will do some operations like Add Products, View All Products with reviews, View All Early Product's reviews, View All Purchased Transactions. End User: In this module, there are n numbers of users are present.User should register before doing any operations.Once user registers, their details will best or to the database.After registration successful, he has to login by using authorized user name and password.Once Login is successful user will do some operations like Manage Account, Search Products by keyword and Purchase, View Your Search Transactions, View.